In this category, the crudest measure gave the clearest result: in the 'cold' sectors, more spikes are generated towards the pattern than away from it, whereas in the 'hot' quadrants the relation is zero or reversed. This is still true if the heat is permanently switched off (Fig. 24).
A
B
Fig. 24: Mean polarity indices (PI) for the operantly
(A) and the classically (B) trained groups (N=100 flies each).
In both groups one can notice a rise in PI in the 'cold' sectors (dotted
bars) compared to a rather steep decline in the 'hot' sectors (hatched
bars). One exception is period 8 in the classical test group.
In all the control groups, the indices for the respective sectors were indiscernible and came very close to 0.1 (Fig. 25). Comparing operant test and control groups at t2, both 'cold' and 'hot' indices were significantly different: 'cold' polarity indices were higher in the test than in the control group (p<0.02) and 'hot' indices were lower (p<0.002). Both 'hot' and 'cold' indices show significant correlations with the respective preference indices (Table 5). As can be expected from the low absolute differences in the polarity indices (Fig. 24) these correlations are lower than those for the polarity independent variables (Table 5). Furthermore, the unproportionally large training values commonly observed in the polarity independent variables are missing in the polarity indices, indicating that this parameter is not modulated in a responsive way.
Fig. 25: The mean polarity indices (PI) for both control
groups. Open symbols indicate the operant group, filled symbols the classical
group. Lines were drawn for illustrational purposes only.
As the spike polarity effects are already quite weak in the standard experiment, they diminish even more after classical conditioning. Evaluating the polarity indices for the 'hot' and the 'cold' sectors (Fig. 24B) still shows a steep decline in the polarity index for the 'hot' quadrants, the effect in the 'cold' quadrants is less impressive, however. Consequently, only the 'hot' effect is statistically safeguarded against the controls (p<0.02, taking all test periods into account). Nevertheless, as in the standard experiment, the 'hot' and the 'cold' polarity indices are significantly correlated with the preference indices (Table 5).
Comparing the two test groups at t2 yields the same results: the 'hot' effect is not different in both groups, the 'cold' effect, however, is significantly larger in the standard group (p<0.02) even if the t1 values are subtracted. This is also reflected in the difference in fixation as will be discussed below (3.5.4).
In the operant group an even more subtle but nevertheless significant effect can be observed in the polarity-latency indices. (Fig. 26A). The difference between the test and control group is clearly to be seen throughout the experiment and at t2, when the heat is switched off, it is still statistically reliable (p<0.04). As with the polarity indices, unproportionally large training values are absent, indicating a gradual development of directing spike polarity.
A
B
Fig. 26: The mean polarity-latency indices for the operant
test groups (A, dotted bars) indicate, that on average the first
spikes away from the pattern are generated earlier in quadrants where the
pattern orientation is associated with the reinforcer than in the other
sectors (hatched bars - control group). In the classical groups
(B), there was no detectable pattern of modulation of spike polarity-latency
neither in the test group (dotted bars), nor in the control group,
which did not receive any reinforcement at all (hatched bars). Only
in period 8 The flies from the classical test group generated the first
spikes away from the pattern earlier in quadrants where the pattern orientation
is associated with the reinforcer than in the other sectors.
In the classical group an effect of spike polarity latency can only be seen in period 8 (Fig. 26B). All other periods show the same random fluctuations as the control group.
Since the absolute values are so small and the errors are relatively large, it is not surprising that the two test groups do not differ significantly from each other at t2 (p=0.305).
In all four groups no correlation between the polarity latency indices and the preference indices can be detected (Table 5). This behavior seems thus to be acquired independently of the expression of learning.